Before words: reading western astronomical texts in early nineteenth-century Japan

SUMMARY

In 1803, the most prominent Japanese astronomer of his time, Takahashi Yoshitoki, received a newly imported Dutch translation of J. J. Lalande's ‘Astronomie’. He could not read Dutch, yet he dedicated almost a year to a close examination of this massive work, taking notes and contemplating his own astronomical practices. How did he read a book he could not read? Following the clues Yoshitoki left in his notes, we discover that he found meanings not only in words, but also in what are often taken for granted or considered to be auxiliary tools for data manipulation, such as symbols, units, tables, and diagrams. His rendering of these non-verbal textual elements into a familiar format was crucial for Yoshitoki's reading, and constituted the initial step in the process of integrating Lalande's astronomy into Japanese astronomical practices, and the subsequent translation of the text into Japanese.

Acknowledgements

I would like to thank my mother, Olga Zolotarevsky, for generously sharing with me her expertise in French and in mathematics, and helping me with deciphering Lalande's original text and calculations. I am also grateful to my colleagues at Princeton University, Johns Hopkins University and the Max Planck Institute for the History of Science for insightful comments and useful feedback.

Notes

1 In modern Japanese, the last name is mentioned before the first, and when abbreviated only the last name is given. However, in the early modern period, astronomers, like artists and craftsmen, formed lineages, where biological sons followed their father's footsteps, while students or sons-in-law were adopted into the family and assumed their teacher's last name. Consequently, due to the abundance of Takahashis, Shibukawas, and so forth, people in such lineages were conventionally referred to by their first name.

2 For an overview of approaches in translation theory see Lawrence Venuti, Translation Changes Everything. In the East Asian context especially, it is worth mentioning the works of Haun Saussy on cultural mediation in the European exploration of China (Haun Saussy, Great Walls of Discourse (Cambridge, MA, 2001)); Douglas Howland, who stresses the importance of existing conceptual frameworks in translating Western political theories in late nineteenth-century Japan (Douglas Howland, Translating the West (Honolulu, 2002)); and Lidia Liu's edited volume on translation in China, which stresses the process of the reciprocal establishment of meanings (Lidia Liu, ed. Tokens of Exchange (Durham, 1999)). For the role of cultural background in the translation of scientific work see especially Thongchai Winichakul, Siam Mapped (Honolulu, 1994); and Peter M. Engelfriet, Euclid in China (Leiden, 1998). Recently Federico Marcon explored the complexity of the translation process in early nineteenth-century Japanese natural history. Federico Marcon, The Names of Nature: The Development of Natural History in Japan, 1600–1900 (PhD Dissertation, Columbia University, 2007).

3 Although this could certainly be the case when a translator possesses the linguistic and technical knowledge necessary for an understanding of the text. Such was the case of the Jesuits in China, who worked with their Chinese students to find appropriate words to express the meanings they already understood well. See, for example, Engelfreit (note 2); Roger Hart, ‘Translating the Untranslatable’, in Tokens of Exchange, edited by Lidia Liu (Durham, 1999), 45–73; and Mathias Schemmel, ‘Stevin in Chinese’, in Translating Knowledge in Early Modern Low Countries, edited by Cook and Dupré (Berlin, 2012), 369–85.

4 Scott Montgomery referred to the process of translation as ‘laying down vocabulary and patterns of linguistic formation’ in Scott Montgomery, Science in Translation: Movements of Knowledge through Cultures and Time (Chicago, 2000), 232. Similar prioritizing of words is seen in David Wright, Translating Science: The Transmission of Western chemistry into Late Imperial China, 1840-1900 (Leiden, 2000). In Translation & Subjectivity Naoki Sakai has deconstructed the assumption regarding the unity of languages that derives from the act of placing boundaries between ‘national languages’. Yet his work also primarily focuses on words as the main component of language. Naoki Sakai ‘Translation and the Figure of Border: Toward the Apprehension of Translation as a Social Action’, Profession (2010), 25–34.

5 For investigation of the process of reading I found it extremely beneficial to use the methodologies employed by historians of the book, such as Anthony Grafton, Ann Blair, Nicolas Jardine, Adrian Johns, and others.

6 The only exception was the higher officers’ annual trip to the capital for the purpose of meeting with the shogun.

7 See Grant Kohn Goodman, Japan and the Dutch 1600-1853 (Richmond, England, 2000); Kazuo Katagiri, ‘The Rise and Development of Dutch Learning (Rangaku) in Japan’ in Acta Asiatica, 42 (1982), 1–17; Jirō Numata, Western Learning: A Short History of the Study of Western Science in Early Modern Japan (Tokyo, 1992). The utilitarian side of translation is quite obvious in Shigeru Nakayama, A History of Japanese Astronomy: Chinese Background and Western Impact (Cambridge, MA, 1969), and more recently in Harold Cook, Matters of Exchange: Commerce, Medicine, and Science in the Dutch Golden Age (New Haven, CT, 2007). In many of these works any discrepancy between the original and the translations is explained as a failure that derived from either underdeveloped linguistic skills, a deliberate choice driven by ideology, ‘neo Confucian’ dogmatism, or xenophobic politics, denying any possibility for the actor to genuinely perceive contents in a different way than intended. This sentiment was repeated as recently as 2000 by Scott Montgomery in Science in Translation.

8 This portrayal is clearly seen in works of Tsutomu Sugimoto (see, for example, Tsutomu Sugimoto (杉本つとむ), Edo no hon'yakukatachi [Edo Interpreters 江戸の翻訳たち] (Tokyo, 1995), and Nagasaki Tsūji Monogatari: Kotoba to Bunka no hon'yakusha [The Tale of Nagasaki Interpreters: Translators of Words and Culture 長崎通物語 : 言葉と文化の翻訳者] (Tokyo, 1990). But it is also repeated in more recent works, such as Lissa Roberts, ‘Frontier Tales: Tokugawa Japan in Translation’ in Brokered World: Go-Betweens and Global Intelligence, 1770-1820, edited by Simon Schaffer (Sagamore Beach, MA, 2009), 1–47.

9 Harmen Beukers, ‘Dodonaeus in Japanese: Deshima Surgeons as Mediators’ in Dodonaeus in Japan: Translation and the Scientific Mind in the Tokugawa Period (Leuven and Kyoto, 2001), 281–98, 285.

10 The Dutch titles of the books are Naerkeurige beschryving van der Natuur der vievoetige diern, and Het Cruydtboek. Masuzo Ueno, ‘The Western Influence on Natural History in Japan’, Monumenta Nipponica 19, no. 3/4 (1964), 315–39; Kiyoshi Matsuda, ‘Reception and Spread of Dodonæus’ Cruydt-Boeck in Japan’, in Dodonaeus in Japan (note 9), 191–218, 196–198.

11 We need to remember that except for several officers, most of the Dutch residents of Dejima were not educated in the sciences (if educated at all) and hence could hardly be of any help explaining natural history or medical books. Yōzaburō Shirahata, ‘The Development of Japanese Botanical Interest and Dodonaeus’ Role: From Pharmacopoeia to Botany and Horticulture’ in Dodonaeus in Japan (note 9), 263–79, 267–68.

12 Kulmus's Dutch title is Ontleekundige Tafelen, Genpaku's Japanese title is Kaitai Shinsho 解体新書.

13 Sugita Genpaku (Ogata Tomio trans.) Dawn of Western Science in Japan: Rangaku Kotohajime (Tokyo, 1969), 33.

14 Sugita Genpaku (note 13), 33–34.

15 Western sources included books by Thomas Bartholin, Steven Blankaart, Caspar Bartholin, Volcher Coiter, Ambroise Pare, Johann Vesling, Jean Palfyn, Geradus Blasius, Govert Bidloo, Juan Valverda de Hamusco, and William Cowper. Needless to say, the ability to read these books did not surpass the ability to read the words in Anatomical Tables. Chinese sources included Wuli Xiaoshi 物理小識 Minor Knowledge about Things and their Principles; and Mengxi Bitan 夢溪筆談 Dream Pool Essays. Suketoshi Yajima, ‘The European Influence on Physical Sciences in Japan’, in Monumenta Nipponica 19, no. 3/4 (1964), 340–51; Shizu Sakai, ‘Translation and the Origins of Western Science in Japan’, in The Introduction of Modern Science and Technology to Turkey and Japan, edited by Feza Gunergun and Shigehisa Kuriyama (Kyoto, 1996), 137–57, 137.

16 Jūtei kaitai shinsho [A major revision of ‘A New Book of Anatomy’ 重訂解体新書]. Gentaku's superior knowledge of Dutch helped in correcting some details, but the conceptually demanding parts remained as obscure to him as they were to the authors of The New Book of Anatomy and there was no clear example that showed that Gentaku had a better understanding of Western anatomy. Shōsuke Satō (佐藤昌介), Yōgakushi Ronkō [Analysis of the History of Western Studies 洋学史論] (Kyoto, 1993), 3–64.

17 Ōtsuki Gentaku, Rangaku Kaitei [A Guide to Dutch Studies 蘭学階梯], cited in Jirō Numata, Western Learning: A Short History of the Study of Western Science in Early Modern Japan (Tokyo, 1992), 78.

18 Encyclopedia of Astronomical Calculations (Lisuan quanshu 暦算全書 Jap. Rekisan zensho), written by the famous Chinese scholar of Western astronomy Mei Wending 梅文鼎; Chongzhen Calendar (Chongzhen Lishu 崇禎暦書 Jap. Suitei rekisho), written by Jesuit astronomers Giacomo Rho and Adam Schall von Bell; Compendium of Celestial Phenomena (Lixiang kaocheng 暦象考成 Jap. Rekishō kōsei); and History of Instruments Used in Observatory (Lingtai yixiang zhi 靈台儀象志 Jap. Reidai Gishōshi).

19 Benjamin Elman, On their Own Terms: Science in China, 1550-1900 (Cambridge, 2005).

20 Kunten 訓 点.

21 In ‘Stevin in Chinese’, Matthias Schemmel talks about the importance of the correspondence of translated text to the existing ‘image of knowledge’. In this sense, Jesuit astronomical treatises attempted to maintain the image of knowledge while gradually altering specific practices.

22 Following the teaching of Thomas Kuhn, historians of Japanese astronomy, such as Shigeru Nakayama, Numata Jiro, and Sugimoto Tsutomu, focused on the introduction of heliocentric theory as a pinnacle of astronomical development, and hence tended to underestimate the crucial importance of Jesuit treatises in transmitting practice-shaping information. Similar to developments in European historiography of science, historiography of science in Japan also moved beyond the framework of paradigmatic shifts and cultural clashes.

23 Lixiang kaocheng houbian 暦 象考成後編; in Japanese, Rekishō kōsei gohen. Japanese astronomers usually refer to this book just as The Sequel (後編 gohen).

24 Memoirs of Hazama Shigeyori: Hazama Shigeyori 間重新, ‘Senkō taigyō sensei jiseki ryakki’ [Record of Works by My Father 先考大業先生事迹略記], cited in full in appendix to Watanabe Toshio (渡辺敏夫), Tenmonrekigakushijō ni okeru Hazama Shigetomi to sono ikka [Hazama Shigetomi and His House as Seen from Historical Records 天文暦学史上における間重富とその一家] (Tokyo, 1943), 455.

25 The Dutch version of Bleau's Atlas was entitled Tweevoudigh Onderwiis vab de hemelshe en aardsche Globen and Adam's Treatise was translated as Beschrijving van't Maaksel en Gebruik der nieuwe Hemel- en Aarrd-globen. The Japanese title of Motoki's work is Shinsei tenchi nikyū yōhōki (新制天地二球用法記).

26 Tenchi nikyū yōhōki hyōsetsu (天地二球用法記評説).

27 Hazama Shigetomi (間重富), Tenchi nigi yōhōki hyōsetsu [Critique of The Use of Celestial and Terrestrial Globes 天地二球用法記評説] (1798), Gakushiin Archive.

28 Keil's treatise had been translated into Dutch as Inleidinge tot de Waare Natuuren Sterrekunde.

29 Tadashi Yoshida (吉田忠), ‘Butsurigaku, Dandōgaku, Kagaku’ [Physics, Ballistics, and Chemistry 物理学・弾道学・化学] in Bakumatsu no Yōgaku [Western Learning in the Bakumatsu Period 幕末の洋学] edited by Nakayama Shigeru, (Kyoto, 1984), 149–73, 154.

30 Rekisan 暦算. Nathan Sivin discusses at length the variety of possible meanings the character 暦 could assume in English. Sivin also claims that when referring to the yearly chart, we should translate this term as ‘almanac’ rather than ‘calendar’, in order to convey the divinational significance of some of the included events. (Nathan Sivin, Granting the Seasons (New York, 2009), 38–40). However, I disagree with this distinction for the following reasons. First, regardless of the content, the purpose of the calendar as a basis for social coordination and organization is similar. Second, our own modern calendar is not devoid of religious elements. On the other hand, there were many calendars in early modern Japan that were abbreviated and did not include information about auspicious events. Finally, in the eighteenth and the nineteenth centuries the word ‘almanac’ in English simply meant pre-calculation and did not necessarily imply divination of auspicious events (see, for example, nautical almanacs). Consequently, this would have probably been a very appropriate translation before the twentieth century. But in the modern period it came to signify something qualitatively different than simple calculation, and now it often bears a negative connotation of ‘unscientific’, which is the opposite to the meaning that Sivin would like to convey. Consequently, even though the term “almanac” may have been historically correct, I would like to avoid possible misinterpretations and use “calendar” instead.

31 土御門.

32 There is no perfect calendar and even today we periodically add leap seconds to ‘manually’ correct it.

33 Known as Kansei Calendrical Reform.

34 Yoshitoki detailed his troubles with the eclipse calculations in a letter to a fellow astronomer, Hazama Shigetomi, in 1798 (寛政十年). Cited in full in, Hisashi Uehara (上原久), Takahashi Kageyasu no kenkyū, [Study of Takahashi Kageyasu 高橋景保の研究] (Tokyo, 1977), 144.

35 Inō calculated it as 28.2 Japanese ri or 110.85 km in today's terms. Yoshitoki, however, estimated it should have been closer to 27.5 Japanese ri. Cited in Hisashi Uehara, (note 34), 149.

36 The original book Yoshitoki worked with was lost in a fire just a few years after its import, but it was replaced with an identical edition with which Yoshitoki's students continued to work. This book is still preserved at the archive of National Astronomical Observatory of Japan (NAOJ). There are four volumes to the French original, and five volumes in the Dutch translation Yoshitoki was working with. Hereafter, I refer to the French original as Astronomie, and to the Dutch translation as Astronomia.

37 Hotta-Settsu's official status was a ‘Junior Elder Counselor’ (wakadoshiyori), a second rank after ‘Elders’ (rōju), who were the highest appointed officials. The sum paid for the book was 80 ryō. It is very hard to estimate the economic meaning of Edo currency in modern terms, because the prices not only changed throughout the period and differed regionally, but also because people invested in goods and services in a different manner than during the modern period. According to the Japanese Currency Museum run by the Bank of Japan, during the eighteenth century, 1 ryō was equal to 1 koku (石) of rice, which was approximately 150 kg and was considered to be roughly the amount of rice needed to feed a person for 1 year. On the other hand, 1 ryō was enough to hire 23 carpenters for a day of work.

38 Toshio Watanabe (渡辺敏夫), Kinsei Nihon tenmongakushi [History of Early Modern Japanese Astronomy近世日本天文学史] (Tokyo, 1986), 229–34.

39 The original notes consist of eight notebooks. Several chosen paragraphs, including some of the discussion of the section §2686, were reprinted in vol. 65 of Nihon Shisō Taikei and annotated by Shigeru Nakayama and Hideo Hirose. (Nihon Shisō Taikei [Compendium of Japanese Thought 日本思想大系](Tokyo, 1972), vol. 65, 168–70) I benefited greatly from some of Nakayama's and Hirose's annotations. However, they made editorial choices to omit certain sections that I believe to be crucial to this study, and I found myself disagreeing with some of their statements. For this article I compared the reprinted notes with copies of the original notes preserved at the Hazama Bunko Archive, the National Astronomical Observatory of Japan (NAOJ) Archive, and the Japanese Academy of Science (Gakushiin) Archive. The copies at NAOJ and Gakushiin came from the archive of Inō Tadataka—Yoshitoki's student who copied them. It seems that in these copies the text is arranged in an order closer to the original Dutch, spelling mistakes are corrected, and many side notes are incorporated in the main text. The copy at Hazama Bunko, on the other hand, contains numerous corrections, side notes, and misspellings. I cannot be sure that this is the original but I believe that among the existing copies, the one preserved at Hazama Bunko is the earliest.

40 Merriam-Webster defines parallax as ‘the apparent displacement or the difference in apparent direction of an object as seen from two different points not on a straight line with the object; especially: the angular difference in direction of a celestial body as measured from two points on the earth's orbit’.

41 Note that in the 1771 French original, the first accent in dèveloppée is grave.

42 On mechanical problems that stood as a basis for mathematical formulas see Michael S. Mahoney, ‘The Mathematical Realm of Nature’, in Cambridge History of Seventeenth-Century Philosophy, Vol. I, edited by D. E. Garber and others, (NY: Cambridge University Press, 1998) 702–55.

43 Takahashi Yoshitoki, Rarande rekisho kanken [A Brief Glance at Lalande's Book of Astronomy ラランで暦書管見] (1803), Hazama Bunko Archive. Hereafter: Brief Glance.

44 Brief Glance (note 43), vol. 1, in reference to §2680. This part was omitted in Shigeru Nakayama's partial transcription of the passage in Nihon Shisō Taikei (note 39), vol. 65.

45 Brief Glance (note 43), vol. 1 in reference to §2680.

46 Fr.: pole /Dutch: Pool; Fr.: équateur /Dutch: Equator; Fr. meridian/Dutch: Meridiaan.

47 Takahashi Yoshitoki, Seiyōjin Rarande rekishohyō yōhōkai [Explanation of the Rules of Use of Tables in Westerners Lalande Book of Astronomy 西洋人ラランデ暦書表用法解] (1803), Hazama Bunko Archive.

48 The origin of the sequence is in the ancient Chinese Book of Changes where these sequences appear as 10 celestial stems and 12 terrestrial branches. In eighteenth-century Japan, the 12 terrestrial branches, or the 12 animals signs as they are also known, were conventionally used to signify hours, directions, and also a succession of order; the 10 stems were used as conventional symbols for the numbers 1 to 10.

49 Brief Glance (note 43), vol. 1 in reference to §2680.

50 See, for example, Michael Lynch and Steve Woolgar, eds., Representation in Scientific Practice (Cambridge, 1990); Brian Baigrie, Picturing Knowledge (Toronto, 1996); Wolfgang Lafevre, ed., Picturing Machines (Cambridge, 2004); and Shigehisa Kuriyama, ‘Between Mind and Eye’, in Paths to Asian Medical Knowledge, edited by Charles Leslie and Allan Young (Berkeley, 1992), 21–43.

51 Brief Glance (note 43).

52 In their annotation of this passage in Nihon Shisō Taikei (note 39), Shigeru Nakayama and Hirose Hideo mistakenly claim that these numbers differ from the ones provided by Lalande, but are similar to those of Pierre Bouguer, and that it is unclear how Yoshitoki got them (note 39, vol. 65, p. 169). Yet these numbers are actually mentioned later in the text, in §2690, where Lalande explicitly references Bouguer. The difference between these numbers and those mentioned in §2686 stems from the fact that they refer to different lines in fig. 222—in §2686 the numbers refer to lines MG and ED that represented the length of radii that created the evolute, whereas the numbers mentioned by Yoshitoki (which he found in §2690) refer to the longest and the shortest radii of earth's ellipsoid (EC and MC).

53 Brief Glance (note 43), explanation of § 2686.

54 The pages in original are not cited in Nihon Shisō Taikei.

55 Brief Glance (note 43), explanation of §2686.

56 Egbert Buys, Nieuw en volkomen woordenboek van konsten en weetenschappen (Amsterdam, 1769), ‘Aardkloot’, 35.

57 Brief Glance (note 43), explanation of § 2686. Today we do acknowledge the existence of a little bulge around the equator, albeit nothing as large as suggested by Buys's book. In the late eighteenth century, however, such theories were discredited. Even though Buys published his book in 1774, it was a popular encyclopedia and did not necessarily reflect the most recent developments in geodetical theories, but rather the debates that took place earlier in the eighteenth century. For the discussion of these debates see Mary Terrall, The Man Who Flattened the Earth: Maupertuis and the Sciences in the Enlightenment (Chicago, 2002), esp. p. 57, where she mentions the theories that discuss the possibility of a ‘bulge’ on Earth's equator.

58 Takahashi Yoshitoki, Chikyū daenkei sekidō nisshoku hō [The Law of Eclipses in Elliptically (Alongside the Equator) Shaped Earth 地球楕円形赤道日食法] cited in Toshio Watanabe, (note 38), 229.

59 J. J. Lalande (Translated into Dutch by Arnoldus Bastiaan Strabbe, 1778) Astronomia of Sterrekunde, vol. 3. § 2689, 385. National Astronomical Observatory of Japan.

60 Brief Glance (note 43), numerous instances.

61 Bansho wage goyō 蛮書和解御用. The character 蛮 means ‘barbarian’ and derives from 南蛮 or ‘southern barbarians’—the name by which Jesuits were referred to when they first arrived in Japan in the mid-sixteenth century. By the nineteenth century the term became conventional, stripped of the original derogatory connotations. I therefore think that translating the character 蛮 as ‘barbarian’ in this context is extremely misleading. Nevertheless, Japanese scholars of that time were aware of the etymology and changed the character to sound the same but evoke a different association.

62 The office became Bansho shirabesho 蕃書調所 [Office of Investigation of Foreign Books] in 1855, then Kaiseijo 開成所 [Office of Development], and Kaisei Gakko 開成学校 [School of Development].

63 As a part of his Translation of Westerner's Lalande Book of Astronomy Hazama Shigetomi deals with specific phenomena, such as the ellipsoid shape of the earth which he analyses with figure 221. Hazama Shigetomi, Seiyōjin Rarande rekisho honyaku 西洋人刺羅迭暦書翻訳, Undated. Hazama Bunko Archive.